#!/usr/bin/env python3
import argparse
from pathlib import Path
import re
import subprocess
import sys
import urllib.parse

parser = argparse.ArgumentParser(
    description="""
    noRMS for APT-based systems (Debian, Ubuntu, etc).
    Checks to see which installed packages are attached to RMS (i.e. the FSF and GNU).
    """
)
parser.add_argument('--verbose', '-v', help='Print more stuff', action='count', default=0)
args = parser.parse_args()

verbose = args.verbose

PACKAGE = re.compile(r'^(?P<name>[^/ ]+).*$')

# from https://directory.fsf.org/wiki/GNU:
# `new Set([].map.call(document.querySelectorAll('table tr td:nth-child(3)'), x => new URL(x.innerText).host))`
GNU_DOMAINS = set([
    "www.gnu.org",
    "puszcza.gnu.org.ua",
    "gcc.gnu.org",
    "gcompris.net",
    "gmplib.org",
    # "www.gnome.org", # the GNOME foundation signed https://rms-open-letter.github.io/ so they're OK
    "www.gnucash.org",
    "gnunet.org",
    "www.gnuhealth.org",
    "www.list.org",
    "www.midnight-commander.org",
    "mediagoblin.org",
    "www.multiprecision.org",
    "www.mpfr.org",
    "www.nano-editor.org",
    "jemarch.net",
    "www.r-project.org",
    "savannah.gnu.org",
    "www.texmacs.org",
    "unifoundry.com"
])

rms_count = 0

all_packages = subprocess.check_output(['apt', 'list', '--installed'], text=True)
packages = [PACKAGE.match(line)['name'] for line in all_packages.splitlines() if line != 'Listing...']
all_info = subprocess.check_output(['apt', 'show', *packages], text=True)
infos = [x for x in all_info.split('\n\n') if len(x.strip()) > 0]

package_homepage_map = dict()
for info in infos:
    lines = info.splitlines()
    package_lines = [line for line in lines if line.startswith('Package: ')]
    if len(package_lines) == 0:
        print('what uhhhh the fuck?')
        print(info)
        print()
        continue
    package = package_lines[0][len('Package: '):]

    homepage_lines = [line for line in info.splitlines() if line.startswith('Homepage: ')]
    if len(homepage_lines) == 0:
        if verbose >= 1:
            print(f'no webpage found for {package}, skipping')
        continue
    homepage = homepage_lines[0][len('Homepage: '):]

    package_homepage_map[package] = homepage

for package in packages:
    if package not in package_homepage_map:
        continue
    webpage = package_homepage_map[package]
    parsed_url = urllib.parse.urlparse(webpage)
    host = parsed_url.netloc
    if verbose >= 2:
        print('package:', package, 'host:', host, 'webpage:', webpage)
    if host in GNU_DOMAINS:
        print(f'error: {package} is GNU software (website is {webpage})')
        rms_count += 1

print()
print('Total RMS-affiliated packages found:', rms_count)

if rms_count == 0:
    print('Hell yeah!')
    sys.exit(0)
else:
    sys.exit(1)